The Thesis explores additional applications of LAP's Aquarius external laser alignment verification Phantom by examining geometric accuracy of magnetic resonance images commonly used for planning intracranial stereotactic radiation surgery (ICSRS) cases. The scans were performed with MRI protocols used for ICSRS, and head and neck diagnosis, and their images fused to computerized tomographic (CT) images. The geometric distortions (GDs) were measured against the CT in all axial, sagittal, and coronal directions at different levels. Using the Aquarius Phantom, one is able to detect GD in ICSRS planning MRI acquisitions, and align the external LAP patient alignment lasers, by following the LAP QA protocol. GDs up to about 2 mm are observed at the distal regions of the longitudinal axis in the SRS treatment planning MR images. Based on the results, one may recommend the use of the Aquarius Phantom to determine if margins should be included for SRS treatment planning. / by Daniel Vergara. / Thesis (M.S.)--Florida Atlantic University, 2012. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2012. Mode of access: World Wide Web.

Two photon microscopy is one of the fastest growing methods of in-vivo imaging of the brain. It has the capability of imaging structures on the scale of 1μm. At this scale the wavelength of the imaging field (usually near infra-red), is comparable to the size of the structures being imaged, which makes the use of ray optics invalid. A better understanding is needed to predict the result of introducing different media into the light path. We use Wolf's integral, which is capable of fulfilling these needs without the shortcomings of ray optics. We predict the effects of aberrating media introduced into the light path like glass cover-slips and then correct the aberration using the same method. We also create a method to predict aberrations when the interfaces of the media in the light-path are not aligned with the propagation direction of the wavefront. / Includes bibliography. / Dissertation (Ph.D.)--Florida Atlantic University, 2015. / FAU Electronic Theses and Dissertations Collection

Medical images have been used increasingly for diagnosis, treatment planning, monitoring disease processes, and other medical applications. A large variety of medical imaging modalities exists including CT, X-ray, MRI, Ultrasound, etc. Frequently a group of images need to be compared to one another and/or combined for research or cumulative purposes. In many medical studies, multiple images are acquired from subjects at different times or with different imaging modalities. Misalignment inevitably occurs, causing anatomical and/or functional feature shifts within the images. Computerized image registration (alignment) approaches can offer automatic and accurate image alignments without extensive user involvement and provide tools for visualizing combined images. This dissertation focuses on providing automatic image registration strategies. After a through review of existing image registration techniques, we identified two registration strategies that enhance the current field: (1) an automated rigid body and affine registration using voxel similarity measurements based on a sequential hybrid genetic algorithm, and (2) an automated deformable registration approach based upon a linear elastic finite element formulation. Both methods streamlined the registration process. They are completely automatic and require no user intervention. The proposed registration strategies were evaluated with numerous 2D and 3D MR images with a variety of tissue structures, orientations and dimensions. Multiple registration pathways were provided with guidelines for their applications. The sequential genetic algorithm mimics the pathway of an expert manually doing registration. Experiments demonstrated that the sequential genetic algorithm registration provides high alignment accuracy and is reliable for brain tissues. It avoids local minima/maxima traps of conventional optimization techniques, and does not require any preprocessing such as threshold, smoothing, segmentation, or definition of base points or edges. The elastic model was shown to be highly effective to accurately align areas of interest that are automatically extracted from the images, such as brains. Using a finite element method to get the displacement of each element node by applying a boundary mapping, this method provides an accurate image registration with excellent boundary alignment of each pair of slices and consequently align the entire volume automatically. This dissertation presented numerous volume alignments. Surface geometries were created directly from the aligned segmented images using the Multiple Material Marching Cubes algorithm. Using the proposed registration strategies, multiple subjects were aligned to a standard MRI reference, which is aligned to a segmented reference atlas. Consequently, multiple subjects are aligned to the segmented atlas and a full fMRI analysis is possible.

"Fluoroscopic imaging devices for interventional radiology and cardiovascular applications have traditionally used image-intensifiers optically coupled to either charge-coupled devices (CCDs) or video pick-up tubes. While such devices provide image quality sufficient for most clinical applications, there are several limitations, such as loss of resolution in the fringes of the image-intensifier, veiling glare and associated contrast loss, distortion, size, and degradation with time. This work is aimed at overcoming these limitations posed by image-intensifiers, while improving on the image quality. System design parameters related to the development of a high-resolution CCD-based imager are presented. The proposed system uses four 8 x 8-cm three-side buttable CCDs tiled in a seamless fashion to achieve a field of view (FOV) of 16 x 16-cm. Larger FOVs can be achieved by tiling more CCDs in a similar manner. The system employs a thallium-doped cesium iodide (CsI:Tl) scintillator coupled to the CCDs by straight (non-tapering) fiberoptics and can be operated in 78, 156 or 234-microns pixel pitch modes. Design parameters such as quantum efficiency and scintillation yield of CsI:Tl, optical coupling efficiency and estimation of the thickness of fiberoptics to provide reasonable protection to the CCD, linearity, sensitivity, dynamic range, noise characteristics of the CCD, techniques for tiling the CCDs in a seamless fashion, and extending the field of view are addressed. The signal and noise propagation in the imager was modeled as a cascade of linear-systems and used to predict objective image quality parameters such as the spatial frequency-dependent modulation transfer function (MTF), noise power spectrum (NPS) and detective quantum efficiency (DQE). The theoretical predictions were compared with experimental measurements of the MTF, NPS and DQE of a single 8 x 8-cm module coupled to a 450-microns thick CsI:Tl at x-ray beam quality appropriate for cardiovascular fluoroscopy. The measured limiting spatial resolution (10% MTF) was 3.9 cy/mm and 3.6 cy/mm along the two orthogonal axes. The measured DQE(0) was ~0.62 and showed no dependence with incident exposure rate over the range of measurement. The experimental DQE measurements demonstrated good agreement with the theoretical estimate obtained using the parallel-cascaded linear-systems model. The temporal imaging properties were characterized in terms of image lag and showed a first frame image lag of 0.9%. The imager demonstrated the ability to provide images of high and uniform spatial resolution, while preserving and potentially improving on DQE performance at dose levels lower than that currently used in clinical practice. These results provide strong support for potential adaptation of this type of imager for cardiovascular and pediatric angiography."

Retinal hypoxia and inadequate oxygen delivery have been implicated as causal for the development of several eye diseases, including diabetic retinopathy, glaucoma, and retinopathy of prematurity. The imaging of oxygen tension in the retina, generated from a measure of the phosphorescence lifetimes of bolus-injected palladium-porphyrin probes, has been used successfully to study retinal oxygen dynamics in numerous animal models. However, the specific parameters for applying this technique in the mouse have not been thoroughly investigated. The goals of this project were to calibrate a newly-constructed phosphorescence lifetime imaging instrument and data analysis software against known oxygen concentrations, to determine specific parameters for probe excitation and image collection and analysis in the mouse eye, and to assess any damage caused to the eye by the technique using histological analysis. An in vitro system was developed for calibration of the probe and for estimation of power of excitation light and camera settings necessary to produce acceptable oxygen maps. In vivo experiments were then performed, and plots indicating camera settings necessary for producing varying qualities of oxygen maps were constructed. Trypsin digestion of retinal tissue was used in an attempt to assess any damage to experimental subjects, but this histological technique was deemed inadequate for analyzing the capillary structures of the mouse eye. Alternatively, damage was assessed using the instrument itself to calculate changes in oxygen tension during the experimental process. The results of this work will allow the phosphorescence lifetime imaging system to be used in the mouse to study how changes in retinal oxygen tension correlate with the progression of eye diseases where oxygen is implicated, including diabetic retinopathy.

A common approach to segment an MRI dataset is to use a standard atlas to identify different regions of interest. Existing 2D atlases, prepared by freehand tracings of templates, are seldom complete for 3D volume segmentation. Although many of these atlases are prepared in graphics packages like Adobe Illustrator® (AI), which present the geometrical entities based on their mathematical description, the drawings are not numerically robust. This work presents an automatic conversion of graphical atlases suitable for further usage such as creation of a segmented 3D numerical atlas. The system begins with DXF (Drawing Exchange Format) files of individual atlas drawings. The drawing entities are mostly in cubic spline format. Each segment of the spline is reduced to polylines, which reduces the complexity of data. The system merges overlapping nodes and polylines to make the database of the drawing numerically integrated, i.e. each location within the drawing is referred by only one point, each line is uniquely defined by only two nodes, etc. Numerous integrity diagnostics are performed to eliminate duplicate or overlapping lines, extraneous markers, open-ended loops, etc. Numerically intact closed loops are formed using atlas labels as seed points. These loops specify the boundary and tissue type for each area. The final results preserve the original atlas with its 1272 different neuroanatomical regions which are complete, non-overlapping, contiguous sub-areas whose boundaries are composed of unique polylines

"Patient motion is a significant cause of artifacts in SPECT imaging. It is important to be able to detect when a patient undergoing SPECT imaging is stationary, and when significant motion has occurred, in order to selectively apply motion compensation. In our system, optical cameras observe reflective markers on the patient. Subsequent image processing determines the marker positions relative to the SPECT system, calculating patient motion. We use this information to decide how to aggregate detected gamma rays (events) into projection images (frames) for tomographic reconstruction. For the most part, patients are stationary, and all events acquired at a single detector angle are treated as a single frame. When a patient moves, it becomes necessary to split a frame into subframes during each of which the patient is stationary. This thesis presents a method for splitting frames based on hypothesis testing. Two competing hypotheses and probability model are designed. Whether to split frames is based on a Bayesian recursive estimation of the likelihood function. The estimation procedure lends itself to an efficient iterative implementation. We show that the frame splitting algorithm performance is good for a sample SNR. Different motion simulation cases are presented to verify the algorithm performance. This work is expected to improve the accuracy of motion compensation in clinical diagnoses."

Molecular imaging of the human body is beginning to revolutionize drug development, drug delivery targeting, prognostics and diagnostics, and patient screening for clinical trials. The primary clinical tool of molecular imaging is Positron Emission Tomography (PET), which uses radioactively tagged probes (radioligands) for the in vivo quantification of blood flow, metabolism, protein distribution, gene expression and drug target occupancy. While many radioligands are used in human research, only a few have been adopted for clinical use. A major obstacle to translating these tools from bench-to-bedside is that PET images acquired using complex radioligands can not be properly interpreted or quantified without arterial blood sampling during the scan. Arterial blood sampling is an invasive, risky, costly, time consuming and uncomfortable procedure that deters subjects' participation and requires highly specialized medical staff presence and laboratories to run blood analysis. Many approaches have been developed over the years to reduce the number of blood samples for certain classes of radioligands, yet the ultimate goal of zero blood samples has remained illusive. In this dissertation we break this proverbial blood barrier and present for the first time a non-invasive PET quantification framework. To accomplish this, we introduce novel image processing, modeling, and tomographic reconstruction tools. First, we developed dedicated pharmacokinetic modeling, machine learning and optimization framework based on the fusion of Electronic Health Records (EHR) data with dynamic PET brain imaging information. EHR data is used to infer individualized metabolism and clearance rates of the radioligand from the body. This is combined with simultaneous estimation on multiple distinct regions of the PET image. A substantial part of this effort involved curating, and then mining, an extensive database of PET, EHR and arterial blood sampling data. Second, we outline a new tomographic reconstruction and resolution modeling approach that takes into account the scanner point spread function in order to improve the resolution of existing PET data-sets. This technique allows visualization and quantification of structures smaller than previously possible. Recovery of signal from blood vessels and integration with the non-invasive framework is demonstrated. We also show general applicability of this technique for visualization and signal recovery from the raphe, a sub-resolution cluster of nuclei in the brain that were previously not detectible with standard techniques. Our framework can be generalizable to all classes of radioligands, independent of their kinetics and distribution within body. Work presented in this thesis will allow the PET scientific and clinical community to advance towards the ultimate goal of making PET cost-effective and to enable new clinical use cases.

In the past decades, the performance of swept source optical coherence tomography (SS-OCT) has experienced an unprecedented improvement which is mainly driven by the rapidly evolving laser technologies: the state-of-art SS-OCT is now tens of dB more sensitive, six orders of magnitude faster, and seeing ten times deeper than the original version of time domain OCT. Regardless of the abovementioned progress, the phase instability is always considered the biggest weakness of SS-OCT and the mainstream belief often states that the mechanical tuning mechanism of the swept source is to blame. In my study, I first developed a high-speed phase-stable SS-OCT based on a new-generation akinetic laser source, which is electrically tuned in wavelength, in the hope of reducing the phase noise to a shot-noise limited level. The experimental results turned out to be contradicted to the conventional phase noise theory, which inspires my discovery of a completely new interpretation for the phase noise in SS-OCT: I proposed that the timing jitter and scanning variability has to be taken into the consideration in the noise model as multiplicative noises. The theory was later validated by another SS-OCT using a different light source. This study for the first time articulated the phase noise’s origin and composition in the SS-OCT. Although the SS-OCT performs relatively worse in phase stability compared with its spectral-domain counterpart (SD-OCT), it is still valuable since it images at a much faster rate than SD-OCT. Therefore, a better temporal resolution could be achieved, which is particularly attractive in areas such as time lapse imaging. I therefore utilize the system along with other two systems to conduct ex vivo imaging on human tracheobronchial epithelium. It is shown that the SS-OCT system could achieve equally good performance in this task. Moreover, thanks to the higher temporal and temporal frequency resolution, finer structure within the frequency response of the ciliary motion is picked up by our system. During the study of ex vivo ciliary imaging, one of the challenges I was confronted with was the enormous amount of data generated by the SS-OCT, especially when high temporal frequency resolution is required. We thus came up with an idea of applying the compressive sensing (CS) to reduce the data size. Currently, we have demonstrated some preliminary results with using CS on reference k-clock channel compression. In the future, we will apply the same theory to compress the sample channel data, especially or time lapse OCT imaging.

Over the past 25 years of development and innovation, optical coherence tomography (OCT) has successfully fills the gap between the ex vivo high-resolution optical microscopy technologies and in vivo low-resolution medical imaging modalities, including computed tomography (CT), magnetic resonance imaging (MRI), ultrasound (US). Ultrahigh resolution (UHR) OCT categorizes OCT systems with an axial resolution below 3 µm in tissue. With the improved resolution, UHR OCT may impart the knowledge of detailed structures of the tissues that are almost close to what histology may provide. This is how UHR OCT can act as a bridge between radiology and histology. This thesis will present an ultrahigh-resolution (UHR) spectral domain (SD) OCT system that features both high axial resolution and long imaging range, and will demonstrate its applications in human myocardium and breast tissue imaging. The UHR OCT system accommodates a supercontinuum light source, and a home-built spectrometer designed to achieve optimized imaging performance. Specifically, the spectrometer features a customized focusing lenses that are comprised of off-the-shelf optics and a 2k-pixel camera to minimize the cost of the instrument. The system manifests an axial resolution of 2.72 µm and a lateral resolution of 5.52 µm, with a large imaging range of 1.78 mm. The sensitivity of the system is 93 dB with a 6-dB sensitivity fall-off range of 0.89 mm. For human myocardium, currently there is no high-resolution non-destructive real-time imaging modality available for biopsy guidance. As a real-time and non-destructive imaging tool, UHR OCT offers additional benefits compared with standard OCT, which are illustrated by successful delineation of micro-structures such as thin elastic fibers and Purkinje fibers in the endomyocardial side. These structures are otherwise not visible within standard-resolution OCT images. Moreover, by adding the cross-polarization (CP) functionality to the UHR SD system, different types of myocardial tissue can be better delineated through the CP contrast. The functional information provided by CP-OCT may also facilitate automatic tissue classification by using A-line signals. For breast tissue imaging, we show qualitatively and quantitatively that UHR OCT images may enable better visualization of detailed features in different types of breast tissue, including healthy and cancerous ones. UHR OCT images of new breast cancer types such as phyllodes tumor, necrotic tumor and fibrotic focus carcinoma are provided for future references. Features developed from UHR OCT images enable a better yield from relevance vector machine (RVM) based stochastic classification model, compared with that from standard resolution OCT images. UHR OCT shows a great promise for automated classification of different tissue types in human breast tissue based off on UHR OCT images. Lastly, we present our endeavor to miniaturize the UHR OCT system on chip. We explore a chip-based optical frequency comb source that may enable UHR OCT at longer wavelengths to achieve better signal penetration in the future. We characterize the performance of the novel source, including the axial resolution and noise, and show that it holds the promise to be adopted in UHR OCT imaging. In addition, we also demonstrate an on-chip tunable reference arm that allows high-topology high-resolution OCT imaging. The compactness of the devices pave the way to the ultimate miniaturization of OCT system.